
Cell2Sentence applies large language models to single-cell RNA data by turning each cell into a text sentence of gene names. The repository holds the code and tutorials for working with these models.
Web appFree
Step 1: Clone the repository and create a conda environment, or install the package with pip.
Step 2: Prepare your single-cell dataset following the data preparation tutorial.
Step 3: Convert expression data to cell sentences with the rank transformation.
Step 4: Run inference with a pre-trained model or fine-tune one on your data.
Step 5: Use the other tutorials for cell type prediction, embeddings or cell generation.
A space-separated list of gene names ordered by descending expression, which lets a language model process a cell as text.
Pythia-based models and the Gemma-based 2B and 27B models are published on Hugging Face, plus a 1B model.
Yes. The package supports fine-tuning, and a tutorial covers fine-tuning on new datasets and custom prompt templates.
You can optionally install flash-attention, which helps most when generating cells with more than a few hundred genes.
BiomniBiomni is a general-purpose biomedical AI agent from Stanford that carries out research tasks across biomedical fields. It combines LLM reasoning, retrieval-augmented planning and code execution.
Biomni-E1Biomni is an open-source biomedical AI agent from Stanford's SNAP group that carries out research tasks from natural-language instructions. It combines language-model reasoning, retrieval-based planning and code execution.
Arize AXArize AX is an AI-focused web app for developer tools workflows, giving users practical model-powered capabilities through a dedicated product or developer interface.
scFoundationscFoundation is a large pretrained model for single-cell transcriptomics with 100 million parameters. The repository provides weights, code and downstream task examples.